GNOME Bugzilla – Bug 736650
crash when saving file
Last modified: 2019-03-23 20:54:27 UTC
+ Trace 234091
Thread 6 (Thread 0x7fffd69e0700 (LWP 15841))
Thread 1 (Thread 0x7ffff7fba980 (LWP 15833))
3.13.91
Where does it crash? In g_type_check_instance() for the thread with the main? The GObject gdb plugin seems to be buggy, because it's the "saved" signal that is sent, not "changed". And the "saved" signal had an error parameter before, that has been removed. So the problem comes maybe from a plugin which was not updated.
Do you still have the core? What does gdb print? In which thread does it crash? Normally all official gedit plugins should work fine, including those in gedit-plugins.
yes. https://bugzilla.redhat.com/show_bug.cgi?id=1144758 with synctex it's crashing
hm. I've disabled synctex and it's crashing
Created attachment 288730 [details] backtrace I can reproduce on 3.14.0
Created attachment 288731 [details] valgrind log G_SLICE=always-malloc G_DEBUG=gc-friendly valgrind -v --tool=memcheck --leak-check=full --num-callers=40 --log-file=valgrind.log gedit Linux\ today\ and\ tomorrow.tex and saved it
I guess this is the reason: ==1354== 1 errors in context 1 of 4175: ==1354== Invalid read of size 8 ==1354== at 0x7F3B40D: g_type_check_instance (gtype.c:4085) ==1354== by 0x7F3247B: g_signal_emit_by_name (gsignal.c:3391) ==1354== by 0x7F17D34: g_closure_invoke (gclosure.c:768) ==1354== by 0x7F29A51: signal_emit_unlocked_R (gsignal.c:3553) ==1354== by 0x7F32190: g_signal_emit_valist (gsignal.c:3309) ==1354== by 0x7F32909: g_signal_emit_by_name (gsignal.c:3405) ==1354== by 0x4E84434: save_cb (gedit-tab.c:2277) ==1354== by 0x79E03DA: g_task_return_now (gtask.c:1077) ==1354== by 0x79E0A65: g_task_return (gtask.c:1130) ==1354== by 0x54836A9: query_info_cb (gtksourcefilesaver.c:550) ==1354== by 0x79E03DA: g_task_return_now (gtask.c:1077) ==1354== by 0x79E03F8: complete_in_idle_cb (gtask.c:1086) ==1354== by 0x81A2AFA: g_main_dispatch (gmain.c:3111) ==1354== by 0x81A2AFA: g_main_context_dispatch (gmain.c:3710) ==1354== by 0x81A2E97: g_main_context_iterate.isra.29 (gmain.c:3781) ==1354== by 0x81A2F4B: g_main_context_iteration (gmain.c:3842) ==1354== by 0x7A0307B: g_application_run (gapplication.c:2282) ==1354== by 0x400DF9: main (gedit.c:135) ==1354== Address 0x1 is not stack'd, malloc'd or (recently) free'd
Is it 100% reproducible? Does it also happen with all plugins disabled? It's maybe a plugin not updated wrt the GeditDocument::saved signal.
yes. this is 100% reproducible. problem in gedit-code-assistance plugin.
gedit-code-assistance needs a new release, I think the problem is fixed in git master.
*** This bug has been marked as a duplicate of bug 738981 ***